suggestions, I'm open to all possibilities. If you're
looking to integrate with an existing bug tracking system, use the Perl DBI
example in the provided source as a reference.
Cheers.
Tracy Brown
[EMAIL PROTECTED]
___
Info-cvs
Christian,
I'm not sure what it is you have in mind. Presumably you want to execute a
little script of some kind and automatically log into some CVS server?
Here's a little perl to do create a key/value pair with a project name and
it's corresponding CVSROOT string. I don't do this sort of
Here's a little one liner that will give you a list of modules using the
history file - it's probably not a definitive list, but then again I don't
really ever use the history file much:
cvs -q history | nawk '{ if ($7 ~ /^[A-Za-z1-9].*/) print $7 }' | sort -u
If you want to be able to select
..snip..
(They already have
ssh access, at this moment, at a later time I might
consider pserver if
I want to add people to the project that shouldn't have a shell
account...)
No, don't do that. There's no protection in CVS to prevent users from
doing almost arbitrary things (nor
, there is no accountability.
Tracy.
-Original Message-
From: Chris Cameron
To: CVS-II Discussion Mailing List; Tracy Brown
Sent: 5/22/01 4:37 PM
Subject: RE: File permissions [make that directory permissions!]
Without contradicting any of Gregs comments on security, which have been
aired
I am unaware of an out-sourcing firm to perform this for you.
All you need to do is copy the repository to the new location. As far as
configuring your authentication mechanisms (pserver, etc.) refer to the
documentation.
Here's a handy command to do this:
# cd /path/to/cvsroot
# tar -cvplf - .
There is a fairly simple way to do this:
cd workingdir/ (where you have a checked out version of the module)
cvs -Q diff -r SOME_PREVIOUS_TAG -r HEAD (or other branch/tag name) module |
nawk '/^Index/ { print $2 }' file.out
or if you have lots of modules in a project/configuration:
cd
This is a matter of perspective. To a security administrator you are passing
through the firewall on what is probably an unauthorized port and the *code*
that you are downloading may contain viruses, worms, etc. I of course do not
feel this way and chances are your administrator doesn't
great with :peserver:
And here I'm just getting the shop on pserver. Fortunately,
I'm not looking for interactivity, just compliance.
Hope this helps.
Tracy Brown.
___
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info
commits, who
takes care of the merges, who does the conflict resolutions, when you build, how
you build, etc.
Tracy
Brown.
-Original Message-From: Jackie Bilicic
[mailto:[EMAIL PROTECTED]]Sent: Wednesday, May 09, 2001 7:39
AMTo: [EMAIL PROTECTED]Subject: New CVS
User
This is true. Not only that, I have had problems issuing the export command
on my modules when templates are enabled with rcsinfo.
Tracy.
-Original Message-
From: Jerry Nairn [mailto:[EMAIL PROTECTED]]
Sent: Thursday, May 10, 2001 1:29 PM
To: Tracy Brown; [EMAIL PROTECTED]
Subject
]]
Sent: Sunday, February 25, 2001 6:23 AM
To: Tracy Brown
Cc: '[EMAIL PROTECTED]'; Assar Westerlund; Jakob Schlyter;
[EMAIL PROTECTED]
Subject: Re: GSSAPI + CVS
Assar, do you have any comments? From your previous patch submission it
sounds
like you have things working... Jakob?
Tracy
I've never gotten this message before during the make install:
if [ ! -f ./CVSvn.texi ]; then ln -s ./CVSvn.texi . || ln ./CVSvn.texi .
|| cp ./CVSvn.texi . ; else true; fi
makeinfo ./cvs.texinfo -o cvs.info
makeinfo: not found
I've patched my CVS-1.11 code with Assar's GSSAPI patch
I'm running the MIT distribution of Kerberos v5-1.2.1 and I've compiled the
cvs server/client with gssapi support. I believe that my inetd.conf and
services files are in order - they're attempting to run cvs with the gserver
option. I'm timing out when attempting to connect to the cvs server.
I had similar issues compiling CVS 1.11 -- here's what I did to fix it:
I modified the LIBS reference in the src/Makefile.
For example:
-- the configure --
./configure --with-gssapi=/usr/local/krb5 --enable-encrypt --enable-server
...snip...
default place for GSSAPI is /usr/local/krb5
checking
Kerberos enthusiasts -
What is the proper inetd.conf configuration for GSSAPI (gserver) support?
I'm guessing it's this:
2401stream tcp nowait root/usr/local/bin/cvs cvs -f
--allow-root=/cvsroot gserver
for configuring a different temp directory
2401stream tcp nowait
Regarding compiling CVS-1.11 with krb51.2.1 I found that the Makefile was
incorrect with it's referece to the crypto libraries. To fix this I modified
the LIBS reference in the src/Makefile.
For example:
-- the configure --
./configure --with-gssapi=/usr/local/krb5 --enable-encrypt
Hello All -
I'm looking into building CVS with Kerberos support. Docs say that I need V4
- has anyone tried compiling CVS 1.11 with the V5 codebase of Kerberos? All
commentary appreciated. Secondly, I'm running 1.11 and I'm curious to the
stability of the development snapshot...
Cheers.
Tracy.
Thomas, here's how I might do it:
First I would modify the commitinfo file in the CVSROOT and add a line to
execute a script. For example:
ALL /usr/local/cvsroot/CVSROOT/bin/commit-validation.sh $USER
Second I would write a simple hash table (email.hash) with the module name
and the email
c:\cygwin\home\davis\keyneticaset
CVSROOT=:pserver:[EMAIL PROTECTED]:/usr/local/cvs
C:\cygwin\home\davis\keyneticac:\mybin\cvs login
(Logging in to [EMAIL PROTECTED])
CVS.EXE [login aborted]: could not find out home directory
From the docs... This means that you need to set the environment
There are a few steps for getting the template to populate the WinCVS
client. Judging from your note, I'd say that you're missing only the very
last step in that you need to re-checkout the module of interest from the
server. Every time you modify the template (commit-template), you'll need to
The only time I've seen a kernel begin to thrash is when memory is consumed
and the paging space is tapped. You can try increasing your paging space -
though the recommended paging allotment is double (in general) your memory,
which in this case, only leaves you with 32 MB paging. Increasing
I believe that the answer is yes - strictly speaking. However, it might be
more trouble then it's worth. The admin -l command locks a revision of a
file for the duration of the edit, that is, until a commit takes place. If
no revision is given, lock the latest revision on the default branch. This
23 matches
Mail list logo
